Case-Based Lessons on Remote Patient Monitoring in Neurology Using Consumer-Grade Wearables

Abstract

Consumer-grade wearables offer promising opportunities for remote patient monitoring (RPM) in neurological disorders, yet their clinical application remains uncertain. In this exploratory analysis, we draw on prospective observational trials using smartwatches in patients with multiple sclerosis, myasthenia gravis, chronic inflammatory demyelinating polyneuropathy, and migraine, who were monitored for 6 to 24 months. Through detailed clinical case narratives, we illustrate both the potential and the limitations of RPM in neurology. Wearable-generated data successfully captured early, clinically meaningful changes, such as the onset of a myasthenic exacerbation, and supported patient engagement in identifying individual triggers, including for migraine. However, external influences such as holidays, infections, or mobility aid use confounded activity signals, underscoring the importance of contextual interpretation. While wearables can enhance neurological care, their integration into clinical workflows is challenged by limited validation and interpretability. Realising their potential requires robust validation in clinical settings and the development of interoperable RPM platforms supported by close collaboration between clinicians, engineers, and patients.

Plain Language Summary

Smartwatches and fitness trackers are increasingly used to monitor health, but how helpful are they for people with neurological conditions? In this study, we followed patients with multiple sclerosis, myasthenia gravis, chronic inflammatory polyneuropathy, and migraine for up to two years using everyday wearables. By looking at real-life cases, we found that these devices can detect important changes in health, such as early signs of worsening symptoms. They also helped some patients better understand their condition and identify possible triggers, like for migraines. However, the data from wearables can be misleading—for example, during vacations, infections, or when using mobility aids—so it’s important to interpret the results carefully. To truly benefit patients, these tools need to be better tested in clinical settings and integrated into easy-to-use systems that support doctors and patients working together.

Background

Digital health technologies (DHTs), broadly defined as “systems that use computing platforms, connectivity, software, and/or sensors for healthcare and related uses”, ¹ have emerged in various fields of medicine, ² including neurology, by offering innovative solutions to longstanding challenges in patient monitoring and disease management. Applications range from clinical monitoring ³ to integration in clinical trials in neurological disorders^2,4,5 and encompass simple, ubiquitous wearable devices ⁶ as well as more complex systems, including telemedicine applications, ⁷ artificial intelligence-driven diagnostics, ⁸ and even immersive virtual reality treatments. ⁹

Building on these innovations, remote patient monitoring (RPM) enables continuous real-time or near real-time tracking in patients’ everyday lives, ¹⁰ with several programmes already reimbursed for patients with heart failure, ¹¹ hypertension, and diabetes ¹⁰ across the globe. By enabling timely detection of physiological changes, RPM supports early interventions and enhances the patient-physician relationship. ¹² Compared to traditional models reliant on periodic on-site visits, RPM approaches provide more dynamic and responsive care. ¹³ Implementing RPM is especially valuable in managing chronic conditions, where symptom progression can be subtle and highly variable, ¹⁴ allowing for proactive management strategies. Furthermore, RPM greatly expands access to specialised care for patients in remote locations or those with rare diseases who face challenges in regularly visiting expert centres. ¹⁵

The field of neurology presents unique challenges that make RPM particularly relevant. ⁴ Many neurological conditions, including multiple sclerosis (MS), chronic inflammatory demyelinating polyneuropathy (CIDP) and myasthenia gravis (MG), are chronic, progressive, or episodic disorders characterized by fluctuating and often unpredictable symptoms that can lead to deterioration in quality of life and, in some cases, life-threatening exacerbations. Other neurological conditions, like migraine, require frequent treatment adjustments based on long-term symptom monitoring, which periodic in-person visits may inadequately capture. RPM bridges this gap by offering continuous, objective assessments in patients’ home environments.

Wearable devices represent one of the most accessible and widely used sets of DHTs within the RPM framework. ⁴ By collecting diverse physiological data - such as heart rate, activity levels, and sleep patterns - they offer continuous, real-time insight into individual health status, ¹⁶ making them popular among the broader population, but also practicable for clinical trials and RPM.^2,4 Increasingly used among the general population - with adoption rates estimated above one third in the US and Germany^17-19 - consumer-grade wearables already extensively track real-world basic functions in large populations. As such, they present a promising resource for enhancing RPM, particularly in neurology, where many diseases potentially affect motor function.

This work explores how consumer-grade wearable RPM can transform neurological care by highlighting both its potential benefits and the key challenges that must be addressed for successful implementation. Through a set of observational case studies, we examine the feasibility and constraints of using consumer-grade wearables for monitoring both comparably frequent neurological diseases (migraine, MS) and rare neuromuscular disorders (MG, CIDP).

Methods

The exemplary cases were taken from prospective observational trials targeting disease-specific questions and deploying consumer-grade smartwatches in MS (NCT06501950, NCT06937723), MG (NCT06441825), CIDP (NCT05723848) and migraine (NCT06862544). Patients were treated at the University Hospital Düsseldorf or cooperating clinics and were equipped with a Withings ScanWatch, model 1 or 2 (Withings, Paris, France).

Wearable data were continuously collected from participants’ Withings ScanWatch and synchronized via the Withings Health Mate app. Participants’ accounts were linked to a secure study web application (built in Python using Dash, version 2.6.1, with data stored in a PostgreSQL database), enabling automatic data transfer and programmatic access through the Withings API. All data were stored on a GDPR-compliant European cloud for secure and reproducible retrieval of step counts and sleep metrics.

For activity analyses, a valid day was defined as any day with reliable step and heart rate measures. Adherence was monitored, with activity data considered adherent if the smartwatch was worn for at least 75% of wake hours (6:00 A.M. to 11:00 P.M.). ²⁰ Based on participants’ reported sleep patterns, a night cycle was defined as 10:00 P.M. to 8:00 A.M., and sleep adherence was defined as wearing the smartwatch for at least 75% of this period. ²¹ Daily step count was used as the primary metric. The median daily step count over the displayed period was calculated for each participant. To illustrate short-term trends, a 7-day rolling average was computed using a right-aligned window of 7 consecutive days (ie, the mean of the current day and the 6 preceding days).

Patients were generally clinically and digitally followed for 6 to 24 months, with established disease-specific clinical scores and questionnaires recorded. Scores included the Quantitative Myasthenia Gravis (QMG) score ²² for MG, Expanded Disability Status Scale (EDSS) ²³ for MS, Inflammatory Neuropathy Cause and Treatment (INCAT) disability score ²⁴ for CIDP and Migraine Disability Assessment Score (MIDAS) for migraine. ²⁵ All analyses were conducted using Python version 3.12.4 (Python Software Foundation, Delaware, USA), and figures were created using matplotlib version 3.10.1.

For the exploratory nature of this article, insightful cases across studies were examined to provide informative, anecdotal evidence of the potential opportunities and limitations of RPM in neurology. Thus, this article does not aim to provide statistically robust findings or detailed study analyses; rather, it illustrates the potential and challenges associated with implementing RPM through consumer-grade wearable devices using actual cases.

Case-Based Insights on Consumer-Grade Wearable Monitoring

Case 1: Activity Levels in the Context of Myasthenic Exacerbation

A case involving subject MG01, a patient with MG treated with an acetylcholinesterase inhibitor (pyridostigmine) and immunotherapy (eculizumab), demonstrates the potential of RPM for early detection of life-threatening exacerbations (Figure 1). MG is a rare, autoimmune neuromuscular disorder that is characterised by muscle weakness and fatigability, due to autoantibodies that impair synaptic transmission. ²⁶ MG01 was hospitalised due to a life-threatening myasthenic exacerbation on August 15, 2024, following an exacerbation of myasthenic symptoms, presenting with a substantially higher (thus worse) QMG score of 12 (compared to 6 on July 31, 2024) and both dysphagia and dyspnoea upon admission. Initially, only a subtle worsening was noted, but the exacerbation later led to oxygen-dependent intensive and intermediate care surveillance and treatment with plasmapheresis.

Figure 1.

Smartwatch-recorded step count decrease preceding and during myasthenic exacerbation. The plot shows the daily step count for patient MG01 from June 28, 2024, to October 15, 2024. Daily step counts are represented in yellow, with a dashed yellow line indicating the median step count for the period. The green line represents the 7-day rolling average, while the red block marks the hospitalisation period

Retrospective analysis of wearable data revealed a notable decline in daily physical activity preceding hospitalisation, as step counts showed a 36% reduction in the prior 3 weeks and a 77% decline in the prior 6 weeks leading up to admission. This suggests that a quantifiable reduction in physical activity may serve as an early indicator of an impending myasthenic exacerbation. Following inpatient treatment, RPM data indicated an improvement in activity levels upon discharge from the hospital on September 11, 2024, corroborating the clinical response to therapy.

This case illustrates how activity-based RPM, through continuous and real-time monitoring, could trigger alarms for earlier clinical intervention, potentially reducing exacerbation severity and hospitalisation rates. Furthermore, with the integration of advanced algorithms, abnormalities can potentially be identified well before patients seek medical advice, thereby enabling proactive interventions that could, in turn, prevent severe, life-threatening episodes and mitigate long-term complications.

Case 2: How Situational Factors Impact RPM Data

The case of CIDP01, a patient with CIDP, an autoimmune disease affecting the peripheral nervous system, leading to muscle weakness and sensory loss, ²⁷ depicts how altered life circumstances can be misinterpreted as a therapy response (Figure 2). A routine check of wearable data revealed a significant increase in step count, initially attributed to a potential therapy change. However, a follow-up phone call to verify the findings and gather the patient’s perspective revealed a different explanation: CIDP01 had been on holiday. Reportedly, the long walk from the hotel to the beach increased motivation and activity level, leading to a median step count of 3500 steps per day during the 3-week holiday, compared to a median of 560 steps per day before the trip.

Figure 2.

Smartwatch-recorded step count increase due to holiday period. The plot shows the daily step count for patient CIDP01 from January 20, 2023, to July 7, 2023. Step counts are shown in yellow, with a dashed yellow line indicating the median step count of 663 for the depicted period. The green line represents the 7-day rolling average, while the red block marks the holiday period. CIDP01 was diagnosed with CIDP in 2019 and treated with immunoglobulins every 4 weeks (INCAT: 2). After patient-reported fluctuations in early 2023, the investigators observed an increase in step count. However, the significant change in activity level can be explained by altered activity levels and motivation due to the patient’s holiday

While changes in physical activity can reflect clinical deterioration or improvement, other (exogenous, non-focal-disease-related) factors can significantly influence the data captured by comparably unspecific consumer-grade wearables, potentially confounding the interpretation of the data. For example, while acute illnesses may temporarily reduce physical activity - reflected in decreased step counts-such reductions do not necessarily indicate a worsening of the chronic neurological condition. Conversely, circumstances such as rehabilitation stays or holiday periods may lead to increased physical activity, which could be misinterpreted as an improvement in health, although this certainly depends on the time horizon of the observation. These variations highlight the challenge of determining the appropriate window of opportunity for intervention, as multiple factors - from lifestyle to seasonality patterns - can influence activity patterns. Combined, these observations and potential confounding circumstances underscore the need to contextualise RPM data within a patient’s lifestyle and situational factors, necessitating daily to weekly assessments of situational factors and leveraging Patient-Reported Outcome Measures (PROMs).

Case 3-5: Impact of Mobility Aids and Comorbidities

For a small number of patients, we aimed to assess the extent to which the use of different mobility aids influences step count and activity levels, considering that most consumer-grade wearables are designed for general use in healthy populations. Figure 3 displays daily step counts over approximately 6 months for Patients MG02, MG03 and MS01, all of whom use a mobility aid to navigate outside their homes. The cases below illustrate how the step detection algorithm of wearable devices can lead to inaccurate step counts in populations requiring mobility aids.

Figure 3.

Smartwatch-recorded step count for patients using mobility aids. Each plot shows daily step count for 1 patient in yellow, with median steps indicated by a dashed yellow line and the green line representing the 7-day rolling average. As absolute step counts are not comparable for patients using mobility aids or affected by tremor, interpretation should focus on within-person trends over time. (A) Daily step count for patient MG02 from June 28, 2024, to January 26, 2025. MG02 was diagnosed with generalised MG in 2022 and presented at study baseline with a QMG score of 21 (significantly higher than the mean of the MG-cohort of 10.8), with a median step count of 3429 steps in the depicted period. These elevated scores remained at this high, but stable level during the 6-month study period. Although MG02 primarily uses a wheelchair, they can walk within their flat. Interestingly, their recorded activity levels, with a median daily step count of 3429, were higher than expected, which might be caused by self-propulsion of the wheelchair. (B) Daily step count for patient MG03 from May 2, 2024, to January 12, 2025. MG03 was diagnosed with generalized MG in 2017, showing a QMG score of 9 at study baseline (Myasthenia Gravis Composite (MGC): 18 and Myasthenia Gravis Activities of Daily Living (MG-ADL): 12). When leaving the house, they use a walker. Median step count during the depicted period was 1166 steps per day. (C) Daily step count for patient MS01 from October 20, 2023, to April 29, 2024. With a median step count of 175 steps in the depicted period. Given their substantial MS and essential tremor, MS01 is fully relying on an electronic wheelchair. Still, median step count for the depicted period is 175 steps

Patient MG02 is a patient who is substantially impaired by MG, necessitating the use of a wheelchair whenever leaving the flat (Figure 3A). The patient reported that self-propelling the wheelchair caused the watch to register distance travelled as steps, most likely due to the repetitive wrist movements during propulsion.

Patient MG03, also diagnosed with MG, relies on a walker to navigate daily life (Figure 3B). Due to the minimal wrist movement required when using the walker, the patient reported that the smartwatch failed to register steps, even during activities such as an approximately 2 km walker-assisted walk to their neurologist.

A third patient, MS01, is fully reliant on an electronic wheelchair, given their substantial MS disease score (EDSS: 7.5) - causing muscle weakness and severe spasticity, coordination issues and sensory disturbances by affecting the central nervous system. ²⁹ Typically, the minimal wrist movement required to operate the wheelchair would result in no noticeable daily step counts, limited to transfers in and out of the wheelchair. However, we observed days with substantially higher step counts (median step count of 175 steps per day), likely attributable to the patients’ essential tremor (Figure 3C). Additionally, in April 2024, the patient mistakenly recorded almost 4000 steps in a day; these could later be attributed to the patient’s clapping at a concert.

Among neurological conditions, many neuroinflammatory, neurodegenerative and neuromuscular diseases lead to distinct gait patterns, tremors, or reliance on mobility aids such as walkers or wheelchairs. Most consumer-grade wearables, designed primarily for a healthy population, lack clinical validation in these populations and under conditions that reflect the realities of mobility for these patients. As a result, even well-studied devices may struggle to capture mobility impairments using their activity algorithms. While some wearables are validated also for use among wheelchair users, ³⁰ most are not.

The 3 cases above emphasise the necessity of situation-specific clinical validation for RPM monitoring devices (eg, as part of the V3 framework) ³¹ and highlight a critical gap in clinical validation for consumer-grade wearables in certain populations. Absolute step counts, for instance, are not interpretable in patients using wheelchairs, walkers, or with prominent tremor, and only within-person changes over time should be interpreted. To be truly effective in supporting research and patient care, situation-specific validation must be transparently researched and published to prevent misinterpretations and/or inaccurate data for entire groups of patients. As a corollary, patients and clinicians must understand the use of assistive mobility devices to avoid misleading assessments of mobility and disease progression. However, while over- and undercounting of steps leads to inaccuracies in absolute step counts, limiting analyses of changes in activity levels to within-person comparisons may still provide valuable insights into a patient’s health status under some circumstances.

Case 6: An Example of Wearables and RPM for Patient Education

In July 2024, patient MI01, diagnosed with migraine, experienced 2 migraine attacks. The data provided by the wearable device reveals insights into the sleep and activity patterns before, during, and after a migraine attack (Figure 4).

Figure 4.

Smartwatch-recorded step count and sleep data in the context of migraine attacks. The plots show step counts and sleep data of Patient MI01, diagnosed with migraine and treated with botulinum toxin as well as metoprolol. They experienced a migraine attack on July 13, 2024, lasting 50 h (intensity 9 out of 10 on numerical rating scale) and an attack on July 24, 2024 (6 out of 10 on numerical rating scale), both marked in red. (A) Step Count and Sleep Duration: Daily step count is plotted in yellow, with a dashed line indicating the median step count of 6855 steps for the depicted period. Daily sleep duration is plotted in turquoise with a dashed line representing the median time asleep for the month of July (7.52 h per night). (B) Total sleep duration, light and deep sleep: Daily total sleep duration is plotted in turquoise, with light sleep duration plotted in light blue and deep sleep duration plotted in dark blue. The ratio of deep to light sleep is plotted in green

Prior to the first attack, the 2 preceding nights were marked by unusually short sleep durations of approximately 5 h. On the day before the attack, daily step counts peaked at 16 000, substantially exceeding the patient’s median of 9000 steps. During the attack, sleep duration remained limited to around 5 h, and sleep quality was poor, as indicated by a very low deep-to-light sleep ratio, suggesting almost no deep sleep. Simultaneously, daily step counts dropped sharply to 2000, reflecting reduced physical activity, likely due to the need for rest. Following the attack, physical activity quickly returned to normal, while sleep duration initially increased to 9 h before stabilising at typical levels. For the second migraine attack, a similar, though slightly less pronounced pattern can be witnessed.

RPM also helps to empower patients by providing not only clinicians but patients themselves with real-time access to health measurements. This continuous data stream may encourage patients to actively engage in their own care by using objective metrics to better understand their condition and provide feedback. Within the scope of our studies, we often see patients independently reporting their wearable data and own observations. For instance, patients have observed a reduction in physical activity during a respiratory infection, noted clinically relevant declines in daily step counts, or recognised that an increased respiratory rate may signal the onset of an infection. Although such self-monitoring can be susceptible to confirmation bias, ³² it also enhances patients’ awareness of both their chronic neurological conditions and their general health status.

Against this backdrop, it is crucial for healthcare providers to offer clear guidance on interpreting wearable data and to integrate these metrics with clinical scores and patient-reported outcomes. By doing so, RPM and wearable technologies can serve as powerful tools for personalised disease monitoring.

Incorporating RPM using wearables and app-based headache diaries into migraine care provides physicians with a comprehensive overview of attacks and allows patients to identify their unique patterns of disease progression and trigger factors. This can ultimately foster a more proactive and collaborative approach to managing their health; however, it requires collaboration between patients and clinicians. Further, such an approach requires multifactorial problem-solving, as migraine triggers are diverse, often complex, and not always straightforward, and the circular nature of migraine attacks remains unclear. While an unusual sleep pattern may contribute to triggering an attack, it could also be a symptom of the migraine itself or have no direct relation at all.

Discussion

Implications for Clinical Care

The cases presented illustrate both encouraging and cautionary examples of using consumer-grade wearables for neurological conditions. For instance, while early declines in activity levels may pre-empt a myasthenic exacerbation, the unspecific nature of such parameters - combined with the lack of rigorous validation - currently limits their clinical integration. Nonetheless, growing evidence points to the value of wearable-recorded data across neurological diseases. For example, recent studies in CIDP have shown that smartwatch-derived step counts are a robust indicator of disease severity, ²⁰ and wearable-based sleep metrics are associated with disease burden. ²¹ Similarly, motion sensors analyzing natural limb movements prove to be a more objective, sensitive, and scalable measure of Amyotrophic Lateral Sclerosis progression. ³³ Together, these findings add to the expanding literature on wearable-derived markers across many neurological diseases.^4,34-37 However, translating these findings into clinical practice remains a crucial task and will require close collaboration between clinicians and engineers.

Consumer-grade wearables are accessible and widely used, providing valuable data on physical activity, sleep, and other basic functions. On the other end of the spectrum, technologically refined and more advanced wearables exist, such as bodysuits designed to track disease progression in neurodegenerative diseases affecting gait.^34,35 While such approaches offer higher sensitivity and specificity than unspecific consumer-grade devices, they often come with higher costs and greater usage burden, making them suitable for research settings or clinical trials, but unlikely to be practicable for widespread use and regular disease management. Similarly, establishing disease-specific digital endpoints, as previously accepted by regulators for a neuromuscular disease, ⁵ is essential but time-consuming and costly. Researchers should thus both aim to validate consumer-grade wearables in different clinical conditions while also building dedicated platforms for both data transmission of consumer-grade hardware and disease-specific devices. Promising examples include a platform that integrates several DHTs, such as spirometers and wearables, combined with patient-reported outcome measures into an RPM telemedicine platform for enhancing MG disease management ³ or a customizable application to track individual aspects of MS disease activity. ³⁸

A certain level of digital literacy on the part of both clinicians and patients is crucial for the successful implementation of RPM. Some of these barriers can be overcome using (continuing) educational programmes and through the development of user-friendly interfaces. ³⁹ Within our set of studies, most participants, regardless of age, encountered minimal to no technological difficulties, and when issues did arise, they were typically resolved via a single phone call. A recent feasibility and usability study of a telemedicine platform integrating wearable sensors, spirometry and PROMs in MG reported findings consistent with our observations. ⁴⁰ With adherence ranging between 74.3% and 97.9%, and both patients and clinicians rating the system highly usable, these findings support the feasibility of RPM in neuromuscular disease contexts and underscore the importance of user-friendly interfaces.

While in-person visits remain essential for comprehensive evaluations, their frequency may be adapted to the clinical necessity, saving both time and energy for patients and easing resource demands and costs for healthcare providers. Another common argument for RPM is the potential healthcare cost savings through reduced hospitalisations. However, recent research outside of the neurology setting points out that while RPM can be effective in reducing disease exacerbations and hospitalisations, it may not always result in consistent cost savings across all chronic conditions.^14,41 For instance, in hypertension care, RPM was found to reduce emergency care visits and testing costs, but overall spending increased due to direct RPM reimbursement and more primary care visits. ⁴¹ Thus, designing RPM programmes with higher initial compensation with a gradual decrease over time (eg, in a setting in which most medication adjustments occur within the early months of RPM use) or selectively targeting RPM toward those patients most likely to benefit (such as those with low medication adherence) could enhance cost-effectiveness. Ultimately, for RPM to effectively improve both health outcomes and lead to savings on costs of care, interventions must accurately identify high-risk populations, detect health declines promptly, and provide timely, appropriate responses. ⁴² In the context of neurology and neuroimmunology, RPM may serve as a therapeutic response marker or guide treatment frequency for costly immunotherapies, ⁴³ for example intravenous immunoglobulin (IVIG) treatment in CIDP or neonatal Fc receptor (FcRn) antagonism in generalised MG. Furthermore, in MS, RPM may enable the early detection of clinical progression independent of relapses and support the timely identification of patients eligible for emerging treatment strategies, including Bruton’s tyrosine kinase inhibitors. ⁴⁴

Building Robust RPM Ecosystems

An RPM solution is only as effective as the collective ecosystem that supports it and in which it operates. While an RPM system must be clinically validated and accurately capture physiological data to detect clinical deterioration, its success depends on a robust, well-designed infrastructure. This includes user-friendly interfaces, secure hosting solutions, and (ideally) seamless integration of recorded information into electronic health records. Equally important are the non-technical elements: clear clinical guidelines for physicians on how to respond to RPM alerts or disease exacerbations, protocols that enable the prompt scheduling of in-person visits when necessary, and strategies to overcome systemic challenges within the healthcare system. These include accessible technology for both clinicians and patients, seamless integration into clinical workflows, adequate reimbursement and responsibility models, administrative support, and ensuring health equity, among other requirements critical to successful implementation in clinical management. ⁴⁵

Conclusion

Consumer-grade wearables hold significant potential for enhancing RPM in neurological disorders by capturing early, clinically meaningful changes and supporting patient engagement in self-management. While their clinical integration is currently limited by confounding factors, a lack of clinical validation, and infrastructure challenges, these barriers can be addressed through proper validation, development of interoperable digital platforms, and close collaboration between clinicians, engineers, and patients. When embedded in robust RPM ecosystems, wearables could lead to patient-centric and cost-effective innovations, leveraging technology that many patients already own and use every day. Given the nature of many neurological diseases, as well as typical activity and sleep tracking of wearables, neurology is well-positioned to drive this kind of RPM. ²⁸

Appendix.

Abbreviation Key

ALS

Amyotrophic Lateral Sclerosis

CIDP

Chronic Inflammatory Demyelinating Polyneuropathy

DHT

Digital health technology

EDSS

Expanded Disability Status Scale

FcRN

Neonatal Fc receptor

IVIG

Intravenous immunoglobulin

INCAT

Inflammatory Neuropathy Cause and Treatment Disability Scale

MG

Myasthenia Gravis

MG-ADL

Myasthenia Gravis Activities of Daily Living

MGC

Myasthenia Gravis Composite

MIDAS

Migraine Disability Assessment Score

MS

Multiple Sclerosis

PROM

Patient-reported Outcome Measure

QMG

Quantitative Myasthenia Gravis Score

RPM

Remote patient monitoring

ORCID iDs

Paula Z. Epping https://orcid.org/0009-0002-6219-988X

Marc Pawlitzki https://orcid.org/0000-0003-3080-2277

Ethical Considerations

Ethical approval was obtained for all studies from the Ethics Committee at the University Hospital Düsseldorf: 2022-1881_1 (CIDP) and 2023-2402 (all other studies).

Consent to Participate

All patients provided written informed consent, and all trial activities were conducted in accordance with the Declaration of Helsinki.

Consent for Publication

Written informed consent for both participation and publication were obtained from all patients as part of the general informed consent process.

Data Availability Statement

The data analysed in this study are available from the corresponding author upon reasonable request.*